Free Power Secrets

Making Your Own Fuel

Get Instant Access

Figure 1. Pathways involved in methanogenesis from the bacterial decay of organic matter in marine sediments. Arrow thickness denotes the relative importance of each process. Solid arrows show pathways to methane; dashed arrows show substrates to sulfate reduction.

Large, complex organic polymers are first broken down by hydrolytic and fermentative bacteria to form a range of small molecules known as 'metabolic intermediates'. Some of these metabolic intermediates may be used by other types of bacteria in addition to methanogens. H2 and acetate are 'competitive substrates', and are utilized by whichever organism in the system is the most efficient. For example, sulfate-reducing bacteria have a higher affinity for hydrogen and acetate (Schonheit et al. 1982), than methanogens (Lovley et al. 1994) and thus outcompete them for these substrates.

Thus, in environments where sulfate is present, such as near-surface marine sediments, sulfate reduction is the dominant terminal organic matter mineralization process. In some (rare) cases, when organic matter concentrations are very high, competitive substrates for methanogenesis may be present at high enough concentrations to enable both methanogenesis and sulfate reduction to proceed. Other metabolic intermediates, such as methanol and methylated amines, are used by methanogens alone and are thus 'noncompetitive' substrates. The presence of non-competitive substrates means methanogenesis can occur in environments dominated by other processes, such as sulfate reduction.

The relative flow of organic carbon through the degradation pathways is suggested in Fig. 1 by the thickness of the arrows. The most important substrates for bacterial methanogenesis are acetate and H2:C02. Methane formation from hydrogen and carbon dioxide, or 'carbonate reduction', is an almost universal processes in isolated cultures of methanogens. This has a secondary significance in that these species are autotrophic and do not require complex organic matter for their metabolism. Methane formation from acetate, or 'acetoclastic methanogenesis', is restricted to fewer species, but it is still a significant process in many environments, as discussed below.

Processes which remove hydrogen and thus prevent its accumulation are fundamentally important (Fig. 1). This 'interspecies hydrogen transfer' occurs when a hydrogen-utilizing organism, such as a methanogen (or acetogen), grows syntrophically with a fermentative bacterium. Syntrophic organisms are those which interact as partners to degrade a substance neither can degrade alone. In this case, the methanogen will consume the hydrogen produced by the fermenter and thus maintain a low hydrogen concentration. Hydrogen build-up would inhibit fermentation. If hydrogen levels are kept low by syntrophs, more energy will be produced by fermentation, and more oxidised compounds produced. One such compound is acetate, a key substrate for methanogenic activity. Acetogens can play a critical role as they remove hydrogen and form acetate, and thus increase the importance of acetoclastic methanogens.

Methane oxidation, which typically occurs aerobically (Madigan et al. 2000), can also occur anaerobically (Iversen and Jorgensen 1985). Both produce carbon dioxide from methane. Although anaerobic methane oxidation has been measured using radiotracers (eg. Iversen and Jorgensen 1985; Cragg et al. 1996; Wellsbury et al. 2000), no anaerobic methane-oxidizing bacterium has been isolated. The current perception is that a consortium of syntrophic organisms are involved in anaerobic methane oxidation. These organisms are possibly a methanogen carrying out "reverse methanogenesis" to produce hydrogen (Figure 1), coupled with a sulfate-reducing bacterium utilizing this hydrogen (Hoehler et al. 1994, Hinrichs et al. 1999).


Methanogens are a morphologically diverse group of strictly anaerobic Archaea, which obligately produce methane as an end-product of their metabolism. Archaea are distinct from other bacteria, and many are extremophiles, capable of living in extremes of temperature, pressure, salt, and pH.

3.1. Temperature and pressure

Methanogens can exist at a wide range of temperatures, usually from 4-55°C. Most methanogens are mesophiles, with optimum growth temperatures around 35°C. Methanobacterium frigidum lives in an Antarctic Lake in near-freezing temperatures (Franzmann et al. 1997). Viable methanogens (both H2:C02 and acetate utilizers) have been found in permafrost (Rivkina et al. 1998). At the upper temperature limits, organisms such as Methanopyrus have been isolated from hydrothermal marine sediments. Methanopyrus grows up to 110°C (Stetter, 1996), and other hyperthermophilic methanogens may exist at temperatures up to 113°C, the upper temperature limit for bacteria. Growth at ~110°C is only possible under pressure to prevent water boiling, and hence all bacteria able to grow at these temperatures must also withstand elevated pressure. Other methanogens have been isolated from deep hot marine environments such as oil reservoirs, e. g. M. thermoaggregans (Blotevogel and Fischer, 1985).

Methanogens have been isolated from high salinity environments, including Methanohalophilus, which can grow at salinities up to 4.4M NaCl (Lai and Gunsalus 1992).

Methanogens normally grow in the pH range 6-8, although they are found in acidic peat bogs and in alkaline lakes (to pH 9.7) (Oremland 1988). At lower pH values, acetate methanogenesis may become more significant than H2:C02, because of more efficient competition for H2 by acetogenic bacteria (Phelps and Zeikus 1984).


Methanogenic bacteria, being strict anaerobes, are only active in oxygen-free environments, such as sediments (freshwater and marine), wetlands, the digestive systems of ruminants and termites, and other locations where the respiration of aerobic organisms has removed all traces of oxygen. Methanogenesis is also important in a range of anthropogenic ecosystems such as landfills, anaerobic sewage treatment plants and rice paddies.

4.1. Freshwater environments

In freshwater sediments, rice paddies, swamps and other terrestrial wetland environments, the principal source of carbon for methanogens is acetate, which can account for >70% of the methane produced (e.g. Cappenberg and Prins 1974). Terrestrial environments are the chief source of methane to the atmosphere, and account for -80% of global emissions each year (Crill et al. 1991). By comparison, the oceans and underlying sediments are responsible for only 0.2-3% of the total methane flux, despite their huge surface area.

4.2. Coastal, Marine and Estuarine Sediments

The high concentration of sulfate in seawater allows sulfate-reducing bacteria to outcompete methanogens (Fig. 1). Thus, in anaerobic near-surface sediments, sulfate reduction dominates the terminal oxidation of organic matter, accounting for up to 60% of all organic matter degradation (Jorgensen 1982). Overall, methanogenesis represents only -0.8% of total carbon flow in near-surface estuarine sediments (Wellsbury et al. 1996). However, once the sulfate concentration decreases below -3 mM (Capone & Kiene 1988), significant methanogenesis occurs, and thus methanogenic activity peaks below the zone of sulfate reduction. If non-competitive substrates are present, however, there will be methanogenesis even in the presence of significant quantities of sulfate.


Although the bulk of organic carbon is degraded by bacteria in near-surface sediments, a small amount survives to become deeply buried. Over geological time this accumulates to form the largest global organic carbon reservoir.

Buried organic carbon is resistant to further degradation, enabling only slow methanogenesis. However, direct substrates for methanogenesis, such as H2 and acetate, are also supplied thermogenically at depth, potentially stimulating CH4 production. These processes include thermogenic breakdown of organic matter, deep-sourced fluid flow, or diffuse hydrothermal venting.

5.1. The Deep Marine Biosphere

As depth increases, the temperature and pressure within marine sediments increase, creating conditions more hostile for life. However, as -70% of the Earth's surface is marine, and the sediments below the sea floor can be up to 10 km deep, this represents an enormous potential habitat for anaerobes.

Over the last 15 years, the presence of significant populations of bacteria at depth in marine sediments has been demonstrated (Parkes et al. 2000). Previously, deep marine sediments had been thought too hostile for life, despite the presence of biogenic methane and other indirect evidence for their presence, such as i) chemical changes in pore water; ii) modification of organic compounds; iii) chemical biomarker compounds suggesting the existence of living bacteria; and iv) changes in the stable isotope composition of compounds.

Bacteria have been detected at all sites studied in the Pacific and Atlantic Oceans, and the Mediterranean Sea. Bacterial abundances decrease with increasing depth from the sediment surface, but persist to great depths, generally conforming to the model published by Parkes et al. (1994) (Fig. 2). The current reported maximum depth is 842 mbsf in sediments from the Woodlark Basin, Papua New Guinea, where there were 320,000 cells ml"1 (Taylor et al. 1999). The average depth of sediments beneath the oceans is 500 m, and at this depth, the model predicts a population of 2.76 x 106 cells ml"1. This is only -3% of the typical near-surface abundance, but remains a considerable bacterial population. Indeed, the model indicates that bacteria in deep-sea sediments account for -10% of living biomass on Earth.

Permafrost Methan

Predicted bacterial population (log cells/ml)

Figure 2. Depth distribution of bacterial populations in deep marine sediments. The solid line depicts the general regression line model of Parkes et al. (2000), where log cells = 7.98 - 0.57 log depth. The dashed lines show the 95% prediction limits of the model.

Predicted bacterial population (log cells/ml)

Figure 2. Depth distribution of bacterial populations in deep marine sediments. The solid line depicts the general regression line model of Parkes et al. (2000), where log cells = 7.98 - 0.57 log depth. The dashed lines show the 95% prediction limits of the model.

Bacterial depth distributions in marine sediments vary systematically from the model (Fig. 2) according to the oceanographic setting (Parkes et al. 2000). In high productivity zones such as the Peru Margin, high organic input to the sediments is reflected in relatively large bacterial populations. Conversely, in low productivity zones such as the Eastern Equatorial Pacific, bacterial populations are relatively small. Importantly, both the population size and its activity can actually increase in the deeper sediment layers in situations with fluid and gas flows, such as brine incursion in sediments (eg. the Peru Margin), and deep thermogenic methane (eg. the Japan Sea).

Methane is ubiquitous in deep marine sediments once sulfate reduction has removed sufficient sulfate to enable methanogenesis to dominate anaerobic organic matter degradation. Methanogens have been enriched from depths of 50 m in the Peru Margin (Cragg et al. 1990), although growth was slow, taking 18 months for detectable methane production. Methanogens are notoriously difficult to grow, and maintaining anaerobic conditions whilst sampling for methane over long periods just compounds this problem. Hence, culturing is likely to underestimate or fail to detect methanogens. 14C-labelled substrates are a much more sensitive technique to measure methanogenic activity, and methanogenesis has been detected at a range of deep sediment sites (Parkes et al. 2000), including those with gas hydrates (Fig. 3). At the Blake Ridge hydrate site, low rates of H>:C02 methanogenesis occurred at ca. 700 mbsf, whilst the rate of acetate methanogenesis was much higher (see section 6.2). Specific gene sequences for methanogens have also been detected in deep sediments.

5.2. Oil Reservoirs and Coalfields

Methanogenic bacteria have been isolated from oil reservoirs where significant methanogenesis occurs (Nazina et al. 1995). In many oil formation waters, acetate occurs in high concentrations (Lewan and Fisher 1994). Also, oil reservoir methanogens are able to use unsaturated and monoaromatic hydrocarbons as substrates when growing syntrophically with other anaerobic meso- and thermophiles. A number of potential syntrophic partners have recently been isolated from oil reservoirs, and include Thermotoga elfii (Ravot et al. 1995), T. subterranea (Jeanthon et al. 1995), Desulfacinum infernum (Rees et al. 1995), Desulfotomaculum thermocisternum (Nilsen et al. 1996) and Thermodesulforhabdus norvegicus (Beederetal. 1995).

More recently, a methanogenic consortium has been shown to metabolize saturated hydrocarbons (e.g. hexadecane) to methane in the laboratory (Zengler et al. 1999). The key to the process seems to be the presence of acetogenic bacteria which metabolize hexadecane to acetate and hydrogen, both of which are then utilized by methanogens. The formation of methane from acetate, other volatile fatty acids and hydrocarbons may represent a significant source of methane for hydrates, which are often associated with petroleum deposits.

In some coalfields there is also bacterial methane formation from fossil fuels. Most notable is the San Juan Coalfield, the most prolific coalbed gas basin in the world (Scott et al 1994), where subsurface groundwater flow transports bacteria and probably nutrients through permeable rock strata to the coal beds. The bacteria metabolize hydrocarbons and other organic compounds in the coal to produce secondary biogenic gases, predominantly methane and carbon dioxide. Continuous biogenic gas production is essential to maintain the exceptional gas production of this field. Non-conventional gas resources, such as coalbeds and organic-rich shales, have largely been attributed to thermogenic processes, yet they may contain substantial quantities of biogenic gas.

5.3 Hydrothermal systems

Hydrothermal systems at mid-ocean ridges, subduction zones, back-arc basins and mid-plate hot spots, are a major source of methane to the oceans (Karl, 1995). This methane is thought to be produced by abiological reactions, either degassing of the mantle or high temperature water-rock reactions. During these reactions, substrates for methanogens are also produced (McCollom and Shock 1997) and, therefore, some of the methane may be of bacterial origin (Baross et al. 1997). This suggestion is reinforced by the isolation of very high temperature H2-utilizing methanogens from some hydrothermal systems, including Methanopyrus sp., which grows optimally at 98°C. In sediment-covered hydrothermal systems, such as Guaymas Basin, thermogenic breakdown of organic matter produces a range of low molecular weight organic compounds, including acetate, which may.stimulate methanogenesis (see also section 6.2).

5.4. The terrestrial subsurface

Considerable populations of anaerobic bacteria (106-107 cells/g, Sinclair and Ghiorse 1989), including methanogens, are present in the terrestrial subsurface. Viable methanogens were obtained throughout a depth profile of U. S. Atlantic Coastal Plain sediments down to 300 m (Jones et al. 1989). In addition, sediment slurries produced acetate and methane when incubated anaerobically. Thus deep terrestrial sediments of Cretaceous age still contained metabolizable organic matter. Acetogens can utilize this organic matter to produce volatile fatty acids, substrates for some methanogens (Chapelle and Bradley 1996). Methanogenesis from ancient organic matter also occurs in limestone rocks of the Florida Escarpment (Martens et al. 1991).

The deep crystalline rock aquifers of the Columbia River Basalt Group are 3-5 km thick, and contain only trace amounts of organic carbon. However, high concentrations of CH4 (up to 160 mM) and H2 (up to 60 (iM) are present. Furthermore, the 813C of methane suggests it is biological, consistent with bacterial methanogenesis from H2:C02. Active methanogens were present in the system, and formed the base of a unique deep bacterial ecosystem (Stevens and McKinley 1995). This community is entirely independent of photosynthesis and hence the surface biosphere. Experiments have shown that weathering of these basalt rocks produces H2 for methanogens, although recently questions have been raised about the environmental applicability of these experiments (Anderson et al. 1998). However, this aquifer environment demonstrates the potential for a bacterial ecosystem supported by abiotic processes i.e. H2 generation. If this process is widespread in the Earth's subsurface, it will have significant implications for geosphere processes fuelling the biosphere.

Autotrophic methanogens and homoacetogens, which produce methane and acetate respectively and utilize subterranean hydrogen and bicarbonate, have also been detected to 446 m depth in granitic aquifers in Sweden (Kotelnikova and Pedersen 1997).

5.5. Cold Terrestrial Environments

Methanogenesis in cold environments is an important component of the global methane budget. Wetlands are the largest natural source of methane and about 50% of all wetlands occur in cold high latitude environments. Cold-adapted psycrophilic methanogens have been isolated from an Antarctic Lake (Franzmann et al. 1997), glaciers (Sharp, personal communication) and permafrost sediments (Rivkina et al. 1998). In ice sheets, methane increases with depth, suggesting that methanogens may be utilizing organic matter ground from bedrock by the advancing glacier. A range of different bacteria exist in permanently cold environments, both to create anoxic conditions and to supply substrates for methanogens (Vorobyova et al. 1997, Sharp et al. 1999).


Oceanic hydrate deposits contain predominantly methane, usually -99% of the gas present. Other gases comprising the remaining 1% are generally carbon dioxide, C2-C5 hydrocarbons and hydrogen sulphide. The high ratio of methane to higher hydrocarbons (C1/C2+ ratio), combined with the stable isotope signatures, denotes a bacterial origin for the methane (Kvenvolden 1995).

Two deep ocean sites containing gas hydrates have been the subject of comprehensive microbiological analysis. These sites were located at the Cascadia Margin (Ocean Drilling Program Leg 146) and at Blake Ridge (ODP Leg 164). Analyses included measurements of bacterial numbers and their activity, the presence of different culturable metabolic groups, and estimates of biodiversity using molecular genetic analysis (Marchesi et al. unpublished).

6.1. Microbiology of Hydrate at Cascadia Margin and Blake Ridge

Hydrate was not actually recovered during sampling at Cascadia Margin (Cragg et al. 1995), but its presence was inferred from geochemical and geophysical data, in a discrete zone between 215 and 225 mbsf at Site 889/890. Bacteria populations decreased with increasing depth similar to other sites (Fig. 2). However, the population increased dramatically (x 10) in the discrete hydrate zone. Culturable bacteria, and bacterial activity (methanogenesis from H2:C02, sulfate reduction and methane oxidation) also increased in the same zone. These observations suggest that hydrate stimulated deep sediment bacteria. The high rates of anaerobic methane oxidation indicate that methane is an important substrate for bacteria in this deep habitat. Molecular genetic analysis confirmed the presence of diverse bacterial and methanogenic populations at the site (Marchesi et al. unpublished), and similar microbial diversity has been confirmed in near-surface hydrate deposits (Bidle et al. 1999).

Hydrate was present in all three Blake Ridge sites between -200-450 mbsf, and increased with the strength of the BSR (Paull et al. 1996). Bacteria decreased in number with increasing depth (Wellsbury et al. 2000), similar to the model of Parkes et al. (1994, Fig. 2). However, bacterial populations were significantly stimulated at certain depths within the hydrate zone and directly below the hydrate (Fig. 3), reflecting high concentrations of free gas (Dickens et al. 1997). The number of cells involved in division also increased significantly at these depths at all three sites, suggesting a more active population. A solid sample of methane hydrate recovered from 331 mbsf at Site 997 contained only 2% of the predicted bacterial population for a sediment at that depth. Thus the active bacterial populations associated with hydrate are living in the sediment around the hydrate rather than within the hydrate lattice.

(log cells/ml)

Figure 3. Stimulation of bacterial populations and activity around the BSR at Blake Ridge. The shaded area represents the hydrate zone.

(log cells/ml)

Figure 3. Stimulation of bacterial populations and activity around the BSR at Blake Ridge. The shaded area represents the hydrate zone.

Radiotracer techniques were used to determine bacterial activity at these sites (Parkes et al. 2000). Sulfate reduction was the most important process near the sediment surface, and rates decreased rapidly with depth. Sulfate was removed by -20-30 mbsf. Methanogenesis, methane concentrations, and methane oxidation all increased below this depth. By -100 mbsf, bacterial activity was low, but then increased dramatically around and below 450 mbsf (1.5 to 15 times), associated with the base of the hydrate zone and the free gas beneath (Fig. 3). Deep hydrate deposits are therefore biogeochemically active and this is reflected in increased bacterial growth (thymidine incorporation). Carbon cycling is occurring in this zone through methane, acetate, and carbon dioxide as rates of methanogenesis from H2:C02 and acetate and methane oxidation are all stimulated (Fig. 3).

6.2. Sources of methane in deep marine sediments

Accurately determining the source of methane in hydrates is difficult. Biogenic gas deposits consist predominantly of methane, and are distinct from thermogenic gases, which contain considerable amounts of C2 and higher hydrocarbons. Thus thermogenic gases have a lower C1/C2+ ratio than bacterially produced natural gas.

Stable carbon and hydrogen isotope signatures are also used to distinguish sources of methane. Thermogenic methane tends to be enriched in 13C relative to bacterial methane. Thus, thermogenic methane has a distinctive 513C signature of ca. -20%o to -50%o (Whiticar, 1999). In contrast, bacterially derived methane has a 813C value of between -50%o and -110%o, with methane derived from H2:C02 being more 13C-depleted (-60 to -110%o) than that derived from acetate (-50 to -65%o).

Various environmental factors can reduce the extent of carbon isotope discrimination during bacterial methanogenesis and thus obscure the 813C distinction between different methane sources, e.g., substrate limitation (reservoir effects) and increased temperature. Furthermore, methane oxidation causes an increase in the 13C content of residual methane. All these effects yield methane with a 513C value which more closely resembles thermogenic gas. Despite these factors, the majority of methane in oceanic hydrate is recognized as biogenic, including the hydrate deposits at Blake Ridge and Cascadia Margin.

The hydrogen isotope signature of the methane can also be used to distinguish the particular substrate for bacterial methanogenesis. Methane from acetate has a 8D value more negative than -250%o, whilst H2:C02 methanogenesis generally results in a 8D value of between -150%o and -250%o. This difference is thought to arise because three-quarters of the hydrogen in methane derived from acetate comes from the deuterium-depleted methyl group, and the remaining quarter from water (Whiticar et al. 1986). In H2:C02 methanogenesis, all the hydrogen in methane is ultimately derived from water.

Hydrogen isotope signatures must be used with caution, as bacterial acetate methanogenesis involves hydrogen isotope exchange (de Graaf et al. 1996). In de Graaf s experiments, labelled acetate (CD3COO) was utilized for methanogenesis. As acetate was metabolized, other labelled acetate molecules were formed (CD2HCOO" and CDH2COO") demonstrating that methyl hydrogen atoms exchanged with water, affecting the SD of methane produced. This could also result in the 8D of methane from acetoclastic methanogenesis being more similar to thermogenic or H2:C02 methane.

Methanogenesis from H2:C02 and methane oxidation were measured in the hydrate zone at Cascadia Margin (Table 1). As subsurface hydrate increased, the rate of methane oxidation increased. However, H2:C02 methanogenesis was around 5 orders of magnitude lower than methane oxidation, suggesting that an alternate source must exist at Cascadia Margin to maintain the methane supply. Cragg et al. (1996) suggested there may be gas and fluid flux into the sediment.

Activity rates (nmol/ml/d)









Increasing (—hydrate




























Blake Ridge



















Table 1. Rates of methanogenesis from H2:C02, acetate, and methane oxidation in sediments near the seafloor and at depths associated with hydrate in sediments from Cascadia Margin and Blake Ridge. Gas and fluid venting increases at Cascadia from 888<891 <889/890; Hydrate abundance was similar at Blake Ridge at Sites 994 and 995. Methanogenesis from acetate was only determined at Site 995.

Table 1. Rates of methanogenesis from H2:C02, acetate, and methane oxidation in sediments near the seafloor and at depths associated with hydrate in sediments from Cascadia Margin and Blake Ridge. Gas and fluid venting increases at Cascadia from 888<891 <889/890; Hydrate abundance was similar at Blake Ridge at Sites 994 and 995. Methanogenesis from acetate was only determined at Site 995.

Acetate methanogenesis could be another source of methane for hydrate. The rate of acetate methanogenesis was determined at one of the Blake Ridge Sites (Site 995). Near-surface acetate concentrations were typically low (7|iM at 50 cm depth), indicating that bacterial production and consumption of acetate were closely balanced. In deeper sediment, however, acetate concentrations began to increase and reached -15 mM at 691 mbsf (Wellsbury et al. 1997). This increase was confirmed by three independent methods: isotachophoresis, ion chromatography, and a specific enzymatic technique. High concentrations of acetate at depth stimulated acetoclastic methanogenesis (Table 1; Fig. 3).

Rates of acetate metabolism were sufficiently high that they could not be sustained without a supply of organic carbon into the sediments. Again, fluid flow may be involved as there is evidence for upward migration of high concentrations of dissolved organic carbon (DOC) into the sediments beneath the hydrate zone at this site (Egeberg and Barth 1998)

Importantly, acetate methanogenesis was two orders of magnitude higher than H2:C02 methanogenesis. This high rate of acetate methanogenesis was associated with high concentrations of free gas beneath the BSR. Acetate methanogenesis also exceeded methane oxidation at and below the BSR. Under these conditions, a local supply of bacterial methane to the hydrate is possible, although this is unexpectedly acetate rather than H2:C02.

What supplies acetate to sediments beneath hydrate at Blake Ridge? There is DOC flux, but the mechanism for its generation is unclear. However, as the depth of burial in sediment increases, so does the temperature. Wellsbury et al. (1997) demonstrated that even a small rise in temperature will increase the biological availability of buried organic carbon, before further temperature increases result in thermogenic processes. This newly reactivated organic carbon will allow continued bacterial activity and thus provide substrates for methanogens, including acetate. At Blake Ridge there was a dramatic downcore increase in the Mass Accumulation Rate (Paull et al. 1996) which resulted in the rapid burial of relatively high amounts of organic carbon (up to 1.8%) below -500 mbsf. Thus heating during burial (37°C/km) would have had a particularly marked effect at this site and contributed to the high acetate concentrations. At even higher temperature and greater depth, purely thermogenic processes may also produce low-molecular-weight compounds, including H2. These substrates could migrate upwards to fuel deep methanogenesis.

6.3. Methane oxidation and hydrate

Defining the upper boundary of a gas hydrate stability zone (HSZ) is difficult as there is no clear geophysical feature. However, any methane degassing from the hydrate and subsequently migrating upwards is a potentially significant substrate for methane-oxidizing bacteria. This is demonstrated at both sites on Blake Ridge where methane oxidation was measured using 14C-labelled methane (Wellsbury et al. 2000), as there is a clear maximum in bacterial methane oxidation at 83 mbsf (Fig. 4). The methane oxidation maximum occurs above the zone where hydrate is present, and may reflect the upper boundary of gas hydrate stability. The presence of anaerobic methane-oxidizing bacteria in sediment from above the hydrate zone at Sites 994 and 995 were confirmed in enrichment cultures in the laboratory (Wellsbury et al. 2000). Sterile culture medium was inoculated with sediment and incubated anaerobically under a methane headspace. After 1 year, a decrease in methane and a concomitant increase in carbon dioxide was measured.

Similarly, anaerobic methane oxidation was measured near the seafloor at the Cascadia Margin Site (Table 1), and thus appears to be characteristic of hydrate sites. Methane oxidation would result in highly 13C-depleted carbonate, and the most isotopically depleted 13C found in the marine environment was measured at Cascadia Margin (Elvert et al. 1999).

The capacity of deep sediment bacteria to oxidize methane that has migrated upwards from a gas hydrate deposit is of fundamental significance for global climate. Methane is -21 times more effective as a greenhouse gas than carbon dioxide on a per molecule basis (Andrews et al. 1996). Furthermore, methane release from hydrates could create a positive feedback for global warming; temperature increases would result in hydrate instability and more methane release to the atmosphere. However, if bacterial methane oxidation 'mops up' the slow, continuous release of methane from the upper boundary of the hydrate, then significant quantities of methane will be released to the atmosphere only in catastrophic circumstances, such as sediment slumping events.

anaerobic methane oxidation (nmol/ml/day)

100 150 200

anaerobic methane oxidation (nmol/ml/day)

100 150 200


uepui r



50000 100000 150000 [methane] CyLTkg)


uepui r



50000 100000 150000 [methane] CyLTkg)

Figure 4. Methane oxidation above the Hydrate stability zone (HSZ) at Blake Ridge, site 995. Methane is represented by the solid line; the dashed line with solid circles shows rates of anaerobic methane oxidation measured from 14CH4.


Globally, methane hydrate is estimated to contain around 10,000 Gigatonnes of carbon as CH4 (Kvenvolden 1995), and the bulk of this methane is of bacterial origin. This reservoir contains more than twice the amount of carbon present in global fossil fuel reserves.

The results from Cascadia Margin and Blake Ridge demonstrate that hydrates constitute a unique deep bacterial habitat in marine sediments, as the abundance and activity of bacteria are elevated at depth. Deep fluid and gas flow from the geosphere may play a role in stimulating these bacterial populations. It has been estimated that about 60% of all bacteria on Earth live in sub-seafloor sediments (Whitman et al. 1998). As hydrates are so widespread, their stimulated bacterial populations may substantially increase this estimate, and hence a significant proportion of subsurface marine bacteria may be associated with gas hydrates.


We thank ODP for allowing us to obtain samples, and participate in a number of cruises. This research was funded by grants from the Natural Environment Research Council, UK. We are grateful to Drs Ian Mather, Ed Hornibrook and Fiona Brock for their comments on a draft version of this chapter.

Chapter 9

Movement and Accumulation of Methane in Marine Sediments: Relation to Gas Hydrate Systems

M. Ben Clennell

Universidade Federal da Bahia, Brazil.

Alan Judd

University of Sunderland, United Kingdom

Martin Hovland

Statoil, Stavanger, Norway


Hydrates may occur where thermodynamic conditions permit and where methane concentration in the water exceeds a threshold level, but they will only concentrate where gas flow is focused. Existing models of submarine gas hydrate occurrence encapsulate the system of transport and reactions into a one dimensional model (e.g. Rempel and Buffet 1998, Zatsepina and Buffett 1998, Xu and Ruppel 1999). With this simplification we can constrain key parameters, but it is difficult to capture the geological complexity of real systems. To predict the spatial distribution of hydrates we need to account for the range of mechanisms by which methane can move though the sediments.

Submarine (and lake bottom) gas hydrates form part of a system of methane production, transport, and reaction known as the shallow gas realm. The general characteristics of gases in the first kilometer or so of the sediments are known from geological studies (Kaplan 1974, Floodgate and Judd 1992, etc.) but important questions about the interactions of the biological, chemical and physical components of the system remain to be answered. Clathrate-forming gases, predominantly methane may be produced by bacteria within the hydrate stability zone itself, but generally there is not sufficient labile organic matter deposited with the sediment to produce an appreciable quantity of methane hydrate (Claypool and Kvenvolden 1983, refer to previous chapters). Recycling of biogenic gas at the base of the stability zone is therefore cited as an important process to concentrate hydrates (Paull et al. 1994; Chapter 6). Deeper in the sediments, gases come predominantly from temperature-dependent reactions. These thermogenic gases may migrate long distances vertically and laterally before reaching the hydrate stability zone.

Exactly how hydrocarbons move through sediments, especially those of low-permeability and across structural barriers is a longstanding headache known as the "migration problem" (Schowalter 1979). However, this problem is most commonly studied in the context of petroleum reservoirs and seals. Whilst these generally occur in solid rocks, gas hydrates generally occur in unlithified sediments as the stability zone may be a shallow as the seabed.

In this paper we examine natural gas hydrates in the broader perspective of sediment gas realms (Fig. 1) (cf. Hovland et al. 1995). Porosity, pore size distribution and permeability of the sediments control the rate and modes of transport of the gas and other components. The reaction rates are coupled to the transports, and the proper fluxes must account for sediment compaction (Bonham 1980, Boudreau 1997). The macroscopic and microscopic state of stress in the sediments, and their textures determine the physical distribution of liquid, gas and hydrate phases (Clennell et al. 1999, Henry et al. 1999). To these ingredients we must add geological realism. That is, the inevitable presence of heterogeneity and features such as faults and fractures which disrupt the idealised continuum.

inflow of waler inflow of gas

Figure 1. The gas hydrate system.

inflow of waler inflow of gas

Figure 1. The gas hydrate system.


Judd and Hovland (1992) describe major features of the shallow gas realm, with emphasis on seabed seepages (Hovland, & Judd 1988) and other manifestations of gas migration. Hovland et al. (1994) review geophysically-detectable features of shallow gas in marine sediments, and define the terminology for seismic surveys and sub-bottom profiles. Heggland (1997) and Cooper et al. (1998) highlight the role of high resolution 2D and 3D seismic suveys in identifying gas within the sedimentary section. The role that gas has in shaping the geomorphological evolution of the continental margins is emphasised by Yun et al. (1999) and references therein. From these studies it is evident that shallow gas systems are dynamic: literally in a state of flux, as demonstrated by the existance of active gas seeps.

In rigid rocks it is simple to envisage gas bubbles exisitng within the pore spaces. In soft sediments, however, gas may either expand into existing pore spaces, or it can create its own space by deforming the surrounding matrix; this amounts to an extra degree of freedom in the system (see Clennell et al. 1999). In the first case the gas pressure, Pg> must exceed the water pressure Pw, in the space to be invaded. In the latter case the strength of the sediment must be exceeded by the deforming agent.

Free phase gas in sediments near the sea bed can exist wholly within the original pore space of coarse sediments (sands and gravels), but it is not stable as small bubbles within the very small pores of fine-grained sediments (clays and silts). Rather, in fine-grained sediments gas forms voids of up to a few millimetres across (Wheeler et al. 1990, Abegg et al. 1997) in a matrix of water-saturated sediment (Fig. 3). These gas voids have much lower capillary pressure than interstitial gas. The principles of gas invasion of pore spaces and the formation of gas voids are described in Chapter 2.0.

2.1. The pressure environment.In considering the pressure / stress environment of a marine sediment the sediment and the pore fluids must be considered separately. In normal circumstances, that is when there is no gas present and there is no over or under pressure, the pore water pressure approximates hydrostatic, i.e., the pressure imposed by a continuous column of water extending downwards, through the pore spaces of the sediment, from the sea surface. The overburden pressure equates to the vertical stress imposed by the bulk mass of the overlying sediment. This vertical stress and confining (horizontal) stresses tend to compact the sediment, giving it strength (although fine-grained sediments also posses cohesion as a result of electrochemical forces). However the pore fluid pressure resists compaction.

2.2. Gas saturation and the solubility of methane in pore water

Gas saturation refers to a property of the sediment and means the proportion of the total pore space filled with the free gas phase (bubbles -consisting mainly of methane in the shallow gas realm), the remainder being filled with water. This is unrelated to the amount of methane that is dissolved in the water. Thus any non-zero gas saturation in the pores implies that the coexisting pore water is holding the maximum amount of methane in solution at that P, T. Water with less dissolved methane is undersaturated with respect to CH4. Pore water with more dissolved methane is supersaturated and is out of chemical equilibrium. In this sense, methane saturation is a property of the fluid relating to the relative amount of dissolved CH4.

Solubility of methane is roughly 0.5 to 2.5 x 10"3 mole fraction in water and brines at depths from 300m to 5000 m (Hunt 1997). Pressure, temperature and salinity are sufficient to paramaterize an Equation of State (e.g, Duan et al. 1992) that can predict accurately the solubility of methane in water over a range of conditions. If pressure falls or temperature changes so that the methane saturation of the pore fluid is exceeded, methane gas will exsolve as bubbles (effervescence). A pressure rise can lead to bubbles being reabsorbed and eventually disappearing (evanescence). A certain level of supersaturation is necessary before gas bubbles will nucleate in a liquid, though in the presence of a solid matrix, surface sites for heterogeneous nucleation will be available.

To predict methane solubility in subsea sediments we also have to consider surface interactions and capillarity. As in case of imbibition or drainage, effervescence and evanescence in pores is likely to occur in jumps rather than continuously (Miller 1980). Supersaturation may, however arise in fine grained porous media because of the difficulty of nucleating bubbles of small size and thus high capillary pressure (Claypool 1996). Capillary supersaturation can be significant. Water charged up with oxygen at 10 MPa will after depressurization to atmospheric conditions only exsolve bubbles inside pores of a radius greater than 25 microns (Brereton 1998). This is a supersaturation factor of a few hundred times. Effective pore radii in fine sediments are on order 0.1 to 1 micron, so not only is the capillary entry pressure high, there is also a barrier to exsolution of gas bubbles from the pore fluid.

2.3. Capillary Theory

Gas cannot enter into the porous structure of a sediment or rock until the capillary pressure exceeds a value comensurate with the radius of pore throats that provide access into the interior pore space (Corey 1994). This threshold pressure is also known as the capillary entry pressure. As the saturation of gas increases within a porous medium, the bubbles must distort to fit into the more confined interstices (Fig. 3). Thus we have a functional relationship between saturation and capillary pressure, defining a capillary characteristic curve. The capillary pressure Pc is the pressure difference between immiscible phases, is

Small bubbles can miarale upwaccfs

Larger bubbles are trapped in pewe bodies

Small bubbles can miarale upwaccfs

Larger bubbles are trapped in pewe bodies

Irreducible Water Saturation
Figure 2. Gas voids from Wheeler et al. 1990.

termed the capillary pressure Pc and this is proportional to the surface tension and to the tightness of curvature k of the interface. When the fluid interfaces int-

Irreducible water saturation

0 Water Saturation (%) 100

Drainage curve:

s saturation increases capillary pressure increases

Irreducible water saturation

0 Water Saturation (%) 100

Drainage curve:

s saturation increases capillary pressure increases

Gas Saturation (%)

Imbibition curve:

gas saturation and capillary pressure decrease

Figure 3. Capillary characteristic. Bubbles are trapped in pore bodies as gas saturation declines and water is imbibed. For a given gas saturation the imbibition curve has a lower capillary pressure than the gas invasion / water drainage curve. The residual gas saturation trapped in this way is typically between 10 and 25%. Some water remains in interstices and films even when the gas pressure is high enough to force out most of the mobile liquid. This irreducible water saturation is typically 815%, attaining higher values in clays.

Gas Saturation (%)

-eract with the pore walls (i.e. the sediment grains), the pressure difference is controlled by the relative attraction between the solids and fluids; the more strongly attracted fluid is termed the wetting fluid, the other being non-wetting. The relative degree of wettability changes the angle of contact 0 between the two fluid phases subtended at the pore wall. This factors into the Young-Laplace equation for the capillary pressure Corey (1994):

In marine sediments containing gas and water the contact angle is close to zero and cos q is unity. The interfacial tension of water with methane gas is about 72 mN m"1 at STP, and decreases with temperature (Schowalter 1979, Sachs and Meyn 1995). For a slit-like pore where r is the radius of curvature for the short axis of the slit:

Even though real pore spaces are highly irregular, the uncertainties as to the sizes of pores and the bubbles of gas within them means that we can work with the assumption that pores are cylindrical or slit like.

The internal pressure of gas voids is constrained to lie between the capillary entry pressure of the sediment matrix and the water entry pressure, which is the capillary pressure in a smooth gas bubble inscribed in the cavity. In deformable sediment, the maximum gas pressure is further constrained to be less than the yield stress of the surrounding sediments and greater than the effective confining pressure (Wheeler et al. 1990, Sills et al. 1991). As depth increases matrix stresses force the bubbles to distort and eventually collapse. Little is known about how and over what depth range this occurs, in part because the microscopic stress field in sediments (< 2mm) is much more complicated than the macroscopic continuum scale (> 1 cm).

270 275

g 290 295 300 305

Methane conc. in pore water, mole fraction x 1000

Figure 4. Gas solubility in the hydrate zone from Tohidi et al. (1997). Within the hydrate stability zone, methane is partitioned from the liquid phase into the hydrate phase (Handa 1990). Penetrating further upwards into the hydrate stability zone progressively less methane remains in solution. Thus there is a concentration gradient which exerts a dominant control on methane fluxes and distributions (Rempel and Buffet 1998, Xu and Ruppel 1999).

Was this article helpful?

0 0
Guide to Alternative Fuels

Guide to Alternative Fuels

Your Alternative Fuel Solution for Saving Money, Reducing Oil Dependency, and Helping the Planet. Ethanol is an alternative to gasoline. The use of ethanol has been demonstrated to reduce greenhouse emissions slightly as compared to gasoline. Through this ebook, you are going to learn what you will need to know why choosing an alternative fuel may benefit you and your future.

Get My Free Ebook

Post a comment